Successful aging is a crucial public health challenge. Because accurate trait impressions from faces provide a significant guide to adaptive social interactions, ameliorating any age-related declines in accuracy is integral to addressing that challenge, and this is the long range goal of the proposed research. Accurate assessments of traits like health, hostility, honesty, and competence are important for avoiding contagion and physical harm and seeking appropriate counsel across the lifespan. Yet, there is a dearth of research investigating the accuracy of older adults' (OA) impressions of these traits despite reason to expect age-related decrements. Young adults' (YA) trait impressions are influenced by overgeneralized responses to facial qualities that provide useful information about social interaction possibilities. More specifically, YA tend to generalize their adaptive responses to facial qualities that mark babies, low fitness, or emotion expressions to people whose facial structure merely resembles babies (babyface overgeneralization), low fitness (anomalous face overgeneralization) or an emotion expression (emotion face overgeneralization). Moreover, generalizing from negative facial qualities, including resemblance to an angry face or an anomalous face, fosters accurate trait impressions, whereas generalizing from positive facial qualities, such as high attractiveness or resemblance to a happy face or a baby's face does not. A separate body of research reveals that OA are worse than YA at recognizing negative emotion expressions, including anger, and less responsive than YA to other negatively valenced stimuli, whereas there is little age difference in recognition of happy expressions or response to positively valenced stimuli. The utility of negatively valenced facial cues for achieving accurate impressions coupled with evidence that OA may be less sensitive to such cues motivates the specific aims of the proposed research. Nine studies use behavioral, neural, and psychopharmacological methods to determine whether: 1) OA and YA show more agreement in trait impressions of faces that vary in positivity (medium to positively valenced cues) than those that vary in negativity (medium to negatively valenced cues); 2) OA show less accurate trait impressions than YA; and whether age differences in trait impressions from faces are influenced by 3) decreased dopamine function in OA; 4) greater emotion regulation in OA; 5) more selective processing of personally relevant faces in OA; and 6) more visual dedifferentiation in OA. By integrating research on trait impressions from faces in YA with research on age-related changes in neural function, motivation, and information processing to achieve these specific aims, the proposed research will contribute to understanding when and why impressions change with age as well as what factors contribute to accurate trait impressions at any age. The findings will advance the fields of aging as well as social affective neuroscience, with implications for the development of interventions to reduce age-related vulnerabilities to maladaptive social interactions that can have negative physical, financial, or psychological repercussions.